Chapter Five - Introductory Graph Theory

Definition Let V be a finite non-empty set, and let E ⊆ V × V . The pair

(V, E) is called a directed graph (on V ) or digraph (on V ) where E is the set
of vertices, or nodes, and E is its set of (directed) edges or arcs. We write
G = (V, E) to denote such a graph.
Example Let V = {a, b, c, d, e} with E = {(a, a), (a, b), (a, d), (b, c)}

Example Directed Vs Undirected graph

Let E = {{a, b}, {b, c}, {a, c}, {c, d}}

1
 Definition Let x, y be (not necessarily distinct) vertices in an undirected
graph G = (V, E). An x−y walk in G is a loop-free finite alternating sequence

x = x0 , e1 , x1 , e2 , x2 , e3 , ..., en−1 , xn−1 , en , xn = y

of vertices and edges from G, starting at vertex x and ending at vertex y

involving the n edges ei = {xi−1 , xi }, where 1 ≤ i ≤ n.
The length of this walk is n, the number of edges in the walk.
Any x − y walk where x = y (and n > 1) is called a closed walk. Otherwise,
the walk is called open.
Example 1. {a, b}, {b, d}, {d, c}, {c, e}, {e, d}, {d, b}. This is an a − b walk
of length 6 (edge {b, d} = {d, b} and vertices d and b are repeated).
2. b → c → d → e → c → f a b − f walk of length 5, vertex c is repeated,
but no edge appears more than once
3. {f, c}, {c, e}, {e, d}, {d, a} an f − a walk of length 4, no repetition of either
vertices or edges

Definition Consider any x − y walk in undirected graph G = (V, E).

2
a) If no edge in the x − y walk is repeated, then the walk is an x − y trail.
A closed x − x trail is called a circuit.
b) If no vertex of the x − y walk occurs more than once, then the walk is
called an x − y path.
When x = y, the term cycle is used to describe a closed path.
Definition Let G = (V, E) be an undirected graph. We call G connected if
there is a path between any two distinct vertices of G.
A graph that is not connected is called disconnected.
Example Given undirected graph G on V = {a, b, c, d, e, f, g}, if G is com-
posed of pieces (with vertex sets V1 = {a, b, c, d}, V2 = {e, f, g} and edge sets
E1 = {{a, b}, {a, c}, {a, d}, {b, d}, E2 = {{e, f }, {f, g}} that are themselves
connected, and these pieces are called the (connected) components of the
graph.

Hence, G = (V, E) is disconnected if and only if V can be partitioned into at

least two subsets V1 , V2 such that there is no edge in E of the form {x, y},
where x ∈ V1 and y ∈ V2 .

3
A graph is connected if and only if it has only one component.
Definition For any graph G = (V, E), the number of components of G is
denoted by κ(G).
Definition Let V be a finite non-empty set. We say that the pair (V, E) de-
termines a multigraph G with vertex set V and edge E if for some x, y ∈ V ,
there are two or more edges in E of the form
(a) (x, y) (for directed multigraph) or
(b) {x, y} (for an undirected multigraph).
If G = (V, E) is a graph (directed or undirected), then G1 = (V1 , E1 ) is called
a subgraph of G if ∅ =
̸ V1 ⊆ V and E1 ⊆ E, where each edge in E1 is incident
with vertices in V1 .
Definition Given a (directed or undirected) graph G = (V, E), let G1 =
(V1 , E1 ) be a subgraph of G. If V1 = V , then G1 is called a spanning sub-
graph of G.
Definition Let G = (V, E) be a graph (directed or undirected). If ∅ ̸= U ⊆
V , the subgraph of G induced by U is the subgraph whose vertex is U and
which contains all edges (from G) of either of the form
a) (x, y)∀x, y ∈ U (when G is directed)
or b) {x, y} for x, y ∈ U (when G is undirected).
We denote this subgraph by < U >.
A subgraph G′ of a graph G = (V, E) is called an induced subgragh if there
̸ U ⊆ V , where G′ =< U >.
exists ∅ =

4
 Definition Let V be a set of n vertices. The complete graph on V ,
denoted Kn , is a loop-free undirected graph, where for all a, b ∈ V, a ̸= b,
there is an edge {a, b}.
Definition Let G1 = (V1 , E1 ) and G2 = (V2 , E2 ) be two undirected graphs.
A function f : V1 → V2 is called a graph isomorphism if
a) f is one-to-one and onto and
b) for all a, b ∈ V1 , {a, b} ∈ E1 if and only if {f (a), f (b)} ∈ E2 . When such
a function exists, G1 and G2 are called isomorphic graphs.

Matrix representation of graphs

Adjacency matrix is a square matrix of order (n × n) where n is the num-

ber of vertices in the graph G. Generally denoted by A[aij ] where aij is the
ith row and j th column element.
The general form of adjacency matrix is given as below:

5
  
a11 a12 a13 · · · a1n 
 
a21
 a22 a23 · · · a2n 

 . .. 
 .. ··· ··· . 
 
 
a31 a32 a33 · · · a3n

1; if there is an edge from′ vi′ to′ vj′ ,


where aij =

0
 otherwise

This matrix is termed as adjacency matrix, because an entry stores the

information whether two vertices are adjacent or not.
Note: 1. In adjacency matrix, if the main diagonal elements are zero, the
graph is said to be a simple graph.
2. In case
 of a multigraph
n; n the number of edges from′ vi′ to′ vj′ ,


aij =

0 otherwise


Consider the graph

6
  
0 1 1 0 0 1
 
1 0 1 0 0 0
 
 
1 1 0 1 0 0
 
The adjacency matrix is 



0 0 1 0 1 0
 
 
0 0 0 1 0 0
 
 
1 0 0 0 0 0

Example Find the adjacency matrix for i)

and ii)

Incidence Matrix Let G be a simple undirected graph with m vertices

and n edges, then the incidence matrix I[aij ] is a matrix of order (m × n)
where

1; if vertex i belongs to edges j

aij =

0; otherwise


7
 Consider the gragh G
The incidence matrix of the graph G is of order (5×7). The incidence matrix
relative to the ordering v1 , v2 , v3 , v4 , v5 and e1 , e2 , e3 , e4 , e5 , e6 , e7 is given
below:
 
1 0 0 0 1 0 1
 
 
1 1 0 0 0 1 0
 
 
I= 0 1 1 0 0 0 0

 
0 0 1 1 1 1 0
 
 
0 0 0 1 0 0 1

Path Matrix
Suppose that G is a simple graph  with n-vertices. Then the (n × n) matrix

1; if there is a path fromvi to vj

P = [pij ](n×n) defined by: Pij =

0; otherwise


is known as the path matrix or reachability matrix of the graph G.

Consider the graph G

Therefore, the path matrix of the above graph relative to the ordering v1 , v2 ,

8
v3 , v4 , v5 is given as
 
0 1 0 1 1
 
 
1 0 0 1 1
 
 
P = 0 0 0 0 0 
 
1 1 0 0 1
 
 
1 1 0 1 0

Paths, connectivity, and planar graphs

A path in a graph is a sequence v1 , v2 , ..., vk of vertices each adjacent to the
next, and an edge between each vi to vi+1 so that no edge is chosen more
than once.

Consider the graph G

One possible path is v1 v2 v3 v4 v5 .
Connected Graph A graph (not digraph) G(V, E) is said to be connected
if for every pair of distinct vertices ′ u′ and ′ v ′ in G, there is a path.
A directed graph is said to be strongly connected if for every pair of distinct
vertices ′ u′ and ′ v ′ in G, there is a directed path from ′ u′ to ′ v ′ and also from
′ ′
v to ′ u′ . A directed graph is said to be weakly connected if for every pair
of distinct vertices, there is a path without taking the direction.
Consider the graphs given below:

9
G1 is strongly connected, G2 is weakly connected, G3 is connected, and G4
is disconnected.
Planar Graphs A graph (or multigraph) G is called planar if G can be
drawn in the plane with its edges intersecting only at the vertices of G. Such
a drawing of G is called an embedding of G in the plane.
Look at the graphs below:

(a) is a planar graph.

10
In (b) we have a nonplanar graph; the edges {x, z} and {w, y} overlap at a
point other than a vertex. However, we can redraw the graph as shown in
(c).
k5 is nonplanar.
Theorem If a, b are distinct vertices in a tree T = (V, E), then there is a
unique path that connects these vertices.
Proof Since T is connected, there is at least one path in T that connects
a and b. If there were more, from two such paths some of the edges would
form a cycle. But T has no cycles.

Figure 1: A graph on the left, two possible spanning trees

Minimum Spanning Trees

A Tree is acyclic, undirected, connected graph.

Spanning Trees
Definition For a connected undirected graph G = (V, E), a spanning tree is
a tree T = (V, E ′ ) with E ′ ⊆ E.
Minimum Spanning Trees

11
Suppose we are given a connected, undirected, weighted graph. This is a
graph G = (V, E) together with a function w : ER that assigns a weight to
each edge e.
The minimum (weight) spanning tree (MST) problem is given a connected,
undirected, weighted graph G = (V, E, w), to find spanning tree of minimum
weight, where the weight of a tree T is defined as:
P
w(T ) = e∈E(T ) w(e)
Kruskal’s algorithm
- It works with edges
- Two steps
* Sort edges by increasing edge weight
* Select the first |V | − 1 edges that do not generate a cycle.
Example Use Kruskal’s algorithm to find a minimum spanning tree (MST)

for the following graph:

12
Solution
1. Sort edges by increasing edge weight:
2. Select the first |V | − 1 edges which do not generate a cycle.

Edge Weight Include/Exclude

Edge Weight Include/Exclude
(D, E) 1
(B, E) 4
(D, G) 2
(B, H) 4
(D, C) 3
(A, H) 5
(E, H) 3
(B, F) 6
(G, H) 3
(A, B) 8
(C, F) 3
(A, F) 10
(B, C) 4

Edge Weight Include/Exclude

Edge Weight Include/Exclude
(D, E) 1 ✓
(B, E) 4 ✗
(D, G) 2 ✓
(B, H) 4 ✗
(D, C) 3 ✓
(A, H) 5 ✓
(E, H) 3 ✗
(B, F) 6 Not considered
(G, H) 3 ✓
(A, B) 8 Not considered
(C, F) 3 ✓
(A, F) 10 Not considered
(B, C) 4 ✓

Hence, we get this MST with a total weight of 21.

13
Prim’s algorithm
- It works with nodes (instead of edges)
- Two steps
* Select node with minimum distance
* Update distances of adjacent, unselected nodes
Example Use Prim’s algorithm to find a minimum spanning tree (MST) for

this graph:
Solution
1. Start with any node, say D
2. Update distances of adjacent, unselected nodes
3. Select node with minimum distance
Repeat these steps until all nodes are included
K: Whether in the tree
dv : distance to the tree
pv : Closest node that is in the tree

14
 K dv pv
A
B
C F 3 D
D T 0 -
E F 25 D
F F 18 D
G F 2 D
H

We pick G and update the table for the unselected nodes.

K dv pv
A
B
C F 3 D
D T 0 -
E F 7 G
F F 18 D
G T 2 D
H F 3 G

Next we include C and update the table for the unselected nodes.

Next, we include F. We continue like this until we finally get:

15
 K dv pv
A
B F 4 C
C F 3 D
D T 0 -
E F 7 G
F F 3 C
G T 2 D
H F 3 G

The total weight is 21.

Shortest Path Problem

Suppose we are given a directed weighted graph G = (V, E, w) with node set
V , edge set E, and weight set w. We are given starting node s ∈ V .
The shortest path problem is the problem of determining the shortest path
from node s to all other nodes in the graph.
The weights on the edges are also called costs.

16
Dijkstra’s algorithm
Dijkstra’s algorithm solves shortest path problems for graphs with non-
negative costs, i.e wij ≥ 0 for all (i, j) ∈ E.
Importance of Dijkstra’s algorithm
- Is applied to automatically find directions between physical locations, such
as driving directions on websites like Google Maps.
Steps to implement
Step 1. Initialization
* Assign distance 0 to node s, and label as permanent [The state of node
s is (0, p)]
* Assign to every node a distance value of ∞ and label them as temporary
* Designate the node s as the current node
Step 2. Distance value update and current node designation update
Let i be the index of the current node
(1) Find the set J of nodes with temporary labels that can be reached from
the current node i by an edge (i, j). Update the distance values of these
nodes.
* For each j ∈ J, the distance value dj of node j is updated as follows
new dj = min{dj , dj + wij } where wij is the weight of edge (i, j)
(2) Determine a node j that has the smallest distance value dj among all
nodes j ∈ J, find j ∗ such that minj∈J dj = dj ∗
Step 3. Termination Criterion
If all nodes that can be reached from node s have been permanently labeled,

17
then stop - we are doen.
If we can’t reach any temporary labeled node from the current node, then
all the temporary labels become permanent - we are done.
Otherwise, go to step 2.
Example Use Dijkstra’s algorithm to compute the distance to all nodes from
node 1.

Step 1. Initialization
* Node 1 is designated as the current node
* The state of node 1 is (0, p)
* Every other node has state (∞, t)
Step 2.
* Nodes 2, 3, and 6 can be reached from the current node 1
* Update distance values for these nodes
d2 = min{∞, 0 + 7} = 7
d3 = min{∞, 0 + 9} = 9

18
d6 = min{∞, 0 + 14} = 14
Now among the nodes 2, 3, and 6, node 2 has the smallest distance value.

The status label of node 2 changes to permanent, so its state is (7, p) while
the status of 3 and 6 remains temporary.
* Node 2 becomes the current node.
Step 3
Not all nodes have been reached from node 1, hence we perform another
iteration (back to step 2).
Another implementation of step 2
* Nodes 3 and 4 can be reached from the current node 2
* Update distance values for these nodes
d3 = min{9, 7 + 10} = 9
d4 = min{∞, 7 + 15} = 22
* Node 3 has the smallest distance value

19
* The status label of node 3 changes to permanent, while the status of node
4 remains temporary
* Node 3 becomes the current node
We are not done, so we perform another step 2.
Another step 2
* Nodes 6 and 4 can be reached from the current node 3
* Update distance values for them
d4 = min{22, 9 + 11} = 20
d6 = min{14, 9 + 2} = 11
Node 6 has the smallest distance value. The status label of node 6 changes
to permanent, while the status of 4 remains temporary.
* Node 6 becomes the current node.
We are not done, so we perform another step 2.

Another step 2
* Node 5 can be reached from the current node 6

20
* Update distance value for node 5
d5 = min{∞, 11 + 9} = 20
* Node 5 is the only candidate, so its status changes to permanent and be-
comes the current node.
From node 5 we can’t reach any other node. Hence, node 4 gets permanently
labeled and we are done.

21